Recommendation of a practical guideline for safe tracheostomy during the COVID-19 pandemic

The COVID-19 pandemic is placing unprecedented demands upon critical care services across the world. As of the 10th April 2020, there are currently 70,272 confirmed cases in the UK and over 1.6 million globally, with around half these total cases in Europe [1]. Current evidence from China suggests that between 9.8 and 15.2% of patients will require invasive mechanical ventilation or extracorporeal membrane oxygenation (ECMO) [2‐4]. On 10th April 2020, the intensive care national audit and research centre (ICNARC) published their second report on data from 3883 confirmed COVID-19 admissions to critical care in England, Wales and Northern Ireland. Of these, 1689 were completed episodes with 871 deaths and 818 discharged from critical care, giving a present mortality rate of 51.6% for those requiring critical care admission [5]. As a high consequence infectious disease unit, our institution was one of the first in the UK to treat confirmed COVID-19. As of 4th April 2020, we have had 195 admissions to critical care with a mortality rate of 33.0%.

The insertion of a tracheostomy after around 7–10 days of invasive mechanical ventilation is generally considered a standard of care due to potential to reduce the duration of mechanical ventilation and length of stay on intensive care [6, 7]. The National Confidential Enquiry into Patient Outcome and Death (NCEPOD) study estimated that around 12,000 adult tracheostomies a year were performed in England, Wales and Northern Ireland for such indications in 2014 [8]. Assuming the incidence is the same across Europe (0.2/1,000) then this represents a total of around 100,000 tracheostomies per year. Approximately 70% of these are estimated to be performed by intensivists on intensive care units (ICU) using percutaneous dilatation techniques in the UK, which reflects a shift in practice since the introduction of this technique in 1985 [8].

Genomic analysis of SARS-CoV-2 has determined it to be of a different clade to the betacoronaviruses associated with human severe acute respiratory syndrome (SARS) and Middle East respiratory syndrome (MERS); however, it is well recognised as causing similar respiratory symptoms secondary to a viral pneumonia and in some cases the progression to acute respiratory distress syndrome (ARDS) [4]. The classical radiological appearances are of patchy ground glass opacification and consolidation that may affect multiple lobes and progress to diffuse and dense consolidation. Published studies so far suggest in the early stages of disease there is patchy inflammatory infiltrate, oedema and proteinaceous exudate, with one autopsy study in a patient that died of COVID-19 showing bilateral diffuse alveolar damage with cellular fibromyxoid exudates and interstitial inflammatory infiltrates [9, 10]. The acute lung injury and resultant respiratory failure are responsible for the high levels of invasive mechanical ventilation required for COVID-19 patients.

The COVID-19 pandemic presents unique challenges when considering performing a tracheostomy for such patients for a number of reasons. First, the prognosis of this novel disease is not yet fully understood but mortality rates for those requiring critical care are in the region of 50% [2, 5], which is considerably higher than non-COVID viral pneumonia (22.4%) [5]. This poses a risk of performing futile procedures if a tracheostomy is considered too early. Second, the duration of detectable viral load and correlation with transmission rates during aerosol generating procedures is not yet specifically known. Evidence from China showed infection rates of 3.8% amongst healthcare personnel treating COVID-19 patients [2] but at present, the infection rate related to tracheostomy insertion is unknown. There is no evidence to support a specific technique for tracheostomy insertion in terms of minimising risk of healthcare personnel exposure to airborne droplets.

Whilst post-intubation laryngo-tracheal stenosis is a well-known risk of prolonged endotracheal intubation, there is no evidence it is significantly reduced in patients treated with early tracheostomy (typically less than 10 days). It is too soon to know if there will be any difference amongst the COVID-19 cohort of patients [11, 12]. The incidence of ventilator-associated pneumonia and overall mortality does not improve with early tracheostomy in non-COVID populations [13‐17].

Our institution’s experience is that a significant proportion of patients admitted for mechanical ventilation are weaning and being successfully extubated between days 5 and 10. This is reflected in our median (IQR) length of stay for survivors of 5 (2.7, 11.8) days and supported by the ICNARC report with a median (IQR) duration of advanced respiratory support of 7 (4, 10) days for 1689 confirmed COVID-19 cases. These are mostly younger patients with less co-morbidity [5]. It is also our experience that those who do not get better at this point may have a poor overall prognosis.

Based upon learning from China and Italy, the basic principles of our local strategies and protocols have surrounded early intubation (Fi0₂ > 0.4 with increased work of breathing) for younger patients with less co-morbidity and advanced care planning with early decisions regarding ceiling of care for those least likely to derive benefit. Decision-making around co-morbidity and critical care admission is guided by the clinical frailty score in patients older than 65 as recommended by the National Institute for Health and Care Excellence (NICE), and individualised decisions are made for those aged under 65 [18]. This may reduce the overall number of the elderly and those with multiple co-morbidities that would be more likely to require prolonged mechanical ventilation with little or no derived benefit.

Despite this, there will still be a select cohort of patients who will need longer time to wean and where tracheostomy will be indicated. The benefits can include offering a ‘sealed’ system for ongoing respiratory support. This may be preferable to primary extubation with a high risk of failure, a requirement for high-flow oxygen or non-invasive ventilation. Tracheostomy also allows a lower requirement for sedation thereby facilitating less invasive nursing care; fewer infusion pumps, and the potential for their care to be overseen by non-intensive care trained nursing staff [19]. It is well established that shortening the duration of sedation required minimises risk and duration of associated complications, including delirium, and has been shown to reduce the overall length of stay [20].

A tracheostomy in this cohort of patients brings with it a risk to the healthcare professionals involved both during the actual procedure and in the post-procedure period wherein the patients’ tracheostomy needs to be managed.

In a time where healthcare resources are being placed under a huge strain, there is likely to be a proactive but selective role for tracheostomy. This will benefit both the patient, and provide net benefit for population health resources.

The aim of this paper is to review the current international literature to synthesise a proposed practice model, guideline, and describe the pre-emptive setup and training of a dedicated tracheostomy team.

Most importantly, we outline proposed steps that can be taken to minimise aerosol generation for percutaneous tracheostomy in COVID-19 patients, as we could find no guidance in the existing literature describing this.

We performed a literature review of tracheostomies during the current and previous pandemics consisting of a PubMed search with the terms tracheostomy and Coronavirus, COVID-19, SARS-CoV-2, SARS, MERS. Papers published between 2000 and April 2020 were critically appraised with respect to specific considerations: timing, viral load, staff safety and technique. In addition all publications referencing Coronavirus, COVID-19 and SARS-CoV-2 were screened for relevant content. All COVID-19-related guidance published from oto-rhino-laryngological societies was also reviewed for relevant content.

ENT-UK has also recommended that dedicated teams be established to carry out these procedures with the specific skills and expertise necessary to perform this as safely and in as timely manner as possible [36]. One consideration for such team is having enough personnel to minimise repeated exposure to high levels of aerosol. A study from the SARS outbreak suggested repeated exposure on consecutive shifts can increase risk of healthcare-associated transmission [37]. We could not find any evidence as to the risk related to the number of procedures performed in a single shift.

With safety at the core of our team-based approach; we have initially set out that our tracheostomy team members will not work more than 1 day per week in this role, and not perform more than two tracheostomies per day. Teams will initially be operational twice a week on non-consecutive days with staff on a rotational basis to minimise exposure whilst ensuring consistency in expertise levels across the week. In addition, all team members will be provided with enhanced PPE to include PAPRs.

In our institution, representation from anaesthetics, intensive care (DW/ST) and otorhinolaryngology-head and neck surgery (ORL-HNS) have formed a cross-speciality working group to design the tracheostomy team and develop the best possible guidance using shared expertise. This will also support multi-disciplinary decision-making around timing and appropriateness of tracheostomy. Within the ORL-HNS department, current levels of expertise and experience for both percutaneous and surgical techniques were established for both consultants and senior trainees to help identify appropriate team leaders and ensure adequate skill levels. Cadaveric refresher training was provided over a 1-day session to 25 personnel from the ORL-HNS department. The team had an opportunity to practice the percutaneous technique in line with our action cards twice each. During this day, we were also able to create an instructional video to support learning and ensure our teams are well rehearsed in the COVID-19 percutaneous technique. A link to the edited video is available here: https://​rise.​articulate.​com/​share/​7lfMu5dnIBbNMgtS​4JIZP96GHFdyEWOH​#/​.

Our core tracheostomy team consists of

In the event of an emergency, each team will have a nominated leader identified prior to the procedure. Where possible, it is envisaged that PT will be the technique of choice and performed at the bedside. However, in specific circumstances a ST may need to be performed in an operating theatre.

Based on the current literature and our experience the authors have agreed initial selection criteria for use in our institution. These are set out in Table 1.

Guidelines should remain under constant review as higher level evidence of this novel disease and its outcomes from critical care emerge. They propose a robust but concise list to support decision-making in this challenging area.

To rapidly learn and improve audit and outcomes of practice will be essential and we propose a mandatory minimum audit dataset in Table 2 that can allow local and regional networks to collaborate to improve the knowledge about what will serve this cohort of patients best.

The focus of early post-procedural care is to ensure minimisation of aerosol generation risk to healthcare workers and other patients until any risk has passed. Early measures include keeping the cuff inflated, use of in-line suction, and avoidance of humidified oxygen if possible. Cuff deflation, changing of tracheostomy tube and progress on a decannulation protocol should be deferred until the patient is COVID-19 negative where possible. The provision for nursing and rehabilitation of patients once discharged from the ICU will also need to be considered. The cohorting of patients recovering from COVID-19 with tracheostomies onwards with appropriately trained nursing staff is desirable. A number of other considerations and advice for post-procedural care specific to COVID-19 are detailed in the national tracheostomy safety project and ENT-UK guidance [18, 33].

These proposals provide a robust framework on which to base delivery of tracheostomy services for critical care units during the COVID-19 pandemic, this can prove an invaluable resource as other countries across Europe and the rest of the world could soon be faced with a surge in demand.

At present, there is no high-level evidence beyond case series upon which to make definitive recommendations, but we have based our proposed guidelines upon consensus from the currently available literature to form a pragmatic and safe approach. There is no doubt that as more rapidly emerging higher level evidence becomes available, our recommendations will be refined and improved.

The decision to perform tracheostomy in these patients requires careful consideration, planning and regular scrutiny if it is going to be of net benefit to patients and critical care services in terms of optimising healthcare resource utilisation, ensuring patient and staff safety and providing optimal long-term outcomes.

Urgent planning, training and collaborative data collection will be vital, and ORL-HNS surgeons have the potential to offer a valuable role in supporting critical care teams with this service at a very challenging time.